What is AI Awakening?
Before diving into the book AI觉醒. we’ll take a quick glimpse at "AI Awakening" by Erik Brynjolfsson, the Director of the MIT Initiative on the Digital Economy.
What is AI Awakening?
1. Definition / Concept
According to Erik Brynjolfsson, Director of the MIT Initiative on the Digital Economy, frames AI awakening as the point when AI moves from being powerful but underutilized to creating broad, measurable economic productivity gains.
Quote:
“The technology is real, but the benefits take time to emerge.”
2. Reason for the Delay
The lag in seeing AI’s impact on productivity is due to what Brynjolfsson calls a “restructuring lag”:
“When you have these powerful technologies, you need to make a lot of complementary innovations to get their full potential. It can take decades.”
Analogy: past breakthroughs like the steam engine, electricity, internal combustion engine, and computers also took time before their productivity effects became visible
3. Economic Implications
AI is a general-purpose technology (GPT): it can spur a wave of new innovations, but the full effects depend on adoption across industries.
Even as AI grows the economic pie, benefits may not be evenly shared:
“There’s no guarantee of shared prosperity: incomes haven’t kept pace with economic growth.”
4. Call to Action
Organizations, workers, and policymakers must adapt by reinventing job skills, organizational structures, and measurement systems to capture AI’s benefits.
Brynjolfsson remains cautiously and mindful optimistic:
“The more powerful our tools are, the more we can change the world.”
Now, Let's Dive in Great Read ''AI觉醒:生成式人工智能产业机遇与数字...''
AI 速读
Who are the authors?
Lu Junqun & Li Xuan
1. 鲁俊群Lu Junqun
He is Secretary-General of the 清华大学人工智能国际治理研究院 (Institute for AI International Governance at Tsinghua University).
He is also a researcher at the 北京大学武汉人工智能研究院智能治理研究中心.
His research focus is on AI governance and the digital economy.
He has engaged in significant policy- and governance-related roles: for instance, participating in major international cooperation forums on AI, contributing to governance research on data and social governance in China.
In the context of this book, he brings a strong governance and policy perspective to generative AI and large model issues.
2. 李璇 Li Xuan
She is described as the academic lead of the 清华大学临床医学院数智医健研究中心 (Smart-Digital Health Research Centre) at Tsinghua University.
She serves as Director of the AIGC Special Committee of the 中关村数智人工智能产业联盟 (Zhong guancun Smart-Digital AI Industry Alliance).
She is also an MBA practice instructor at 中国政法大学.
Her background involves driving AI project implementation, digitalization efforts, and training/promotion of talent in the AI and AIGC (AI-Generated Content) field.
2. What is AI Awakening book about?
•This is a professional economic book that deeply explores the
application of cutting-edge new knowledge and ground-breaking practices in
generative AI.
•The book focuses on the new opportunities and challenges for the
industry brought by generative AI, covering core issues behind big models,
various aspects of business transformation, life changes, and governance issues
from an international perspective.
•It details the development and application of generative AI,
stating that its value lies in its powerful generative capabilities, which are
derived from the underlying brain science, data, algorithms, and
computational power.
•The book also explores applications in the arts, allowing machines
to generate new and original content by learning and imitating human creative
patterns and styles.
•It emphasizes the need to pay attention to the ethical and moral
issues of generated AI to ensure its development adheres to ethical and legal
standards, protecting personal privacy and data security.
•Additionally, the book discusses the application of AI companions
in providing emotional support, and analyzes the relationship between the open
source model and technological innovation, noting that the open source paradigm
encourages technology sharing and collaboration.
•Finally, it points out that AI systems learn to better adapt to
human needs through continuous interaction with humans and the receipt of human
feedback.
What are the Technological Innovation and Iterative Progress in light of
the book?
1. The iteration of Technological Innovation and Progress is an ongoing process
involving the development of basic disciplines like mathematics and physics,
and the application of new generation technologies like artificial intelligence.
2. Generative AI:
Is
highlighted as an important innovation direction in the field of AI, capable of
generating creativity and wisdom by learning and imitating what humans create.
3. The continuous upgrading and iteration of core elements such as data and algorithms are the keys to the development of artificial intelligence.
4. AI will continuously integrate, functioning like a massive technological
melting pot.
5. The combination of AI with quantum computing can solve aerospace challenges
that traditional computers cannot handle.
6. The development of deep learning allows computers to analyze vast amounts of
data to learn and recognize images, voices, languages, promoting major
breakthroughs in fields like image recognition, speech recognition, and natural
language processing.
What is the Impact of the overall big model on startups?
1. The emergence of big models provides more opportunities for startups by
allowing them to directly call models without developing algorithms themselves,
and the advent of cloud services lowers market entry barriers.
2. However, computational inefficiencies in large models are a challenge for startups.
3. ChatGPT-type large model products have issues with timeliness, accuracy,
inefficiency, and privacy protection, requiring startups to be more precise and
vertically oriented.
4. The emergence of large models drives the development of artificial
intelligence; for example, the Transformer model was a landmark breakthrough.
5. Key nodes supporting AI development include large-scale high-quality data
sets, strong computing power, breakthroughs in field applications, cooperation
and promotion between research and industry, and the establishment of ethical
and legal frameworks.
What are the values and
Challenges of generative AI in light of the book?
A. Value and Analysis of Generative AI:
1.Tthe value of generative AI lies in its powerful generative
capabilities in generating images, text, or making complex reasoning and
decisions. This capability is supported by brain science, data, algorithms, and
computing power.
2. The application scenario of generative AI is very broad, including nine major
fields such as:
•Business (e.g.,
generating marketing copy)
• Education
(e.g., personalizing learning resources)
• Industrial
design (e.g., assisting in new product design)
•Machinery
manufacturing
• Rural
revitalization
3.The development and application of generative AI face a number of challenges,
including technology regulation, content ecosystem, ethical stewardship,
privacy and security, and digital literacy.
Questions include:
• How to ensure
the authenticity and reliability of the generated content.
•How to protect
the privacy and data security of users.
•How to develop
appropriate ethical guidelines to regulate the behavior of generative
artificial intelligence.
B.Issues and Challenges Facing Generative Artificial Intelligence
The moral and ethical issues of generative AI relate to:
1. Rationality and ethics:
Assessing if dialogue content is accurate,
objective, and adheres to moral and ethical guidelines, and refraining from
generating false or discriminatory content.
2. Data reliability and quality, issues of bias, interpretability and
control:
Generative AI systems must follow ethical
guidelines, guaranteeing fairness of behavior, privacy protection and avoidance
of discrimination.
3.Personal privacy and data security:
Virtual humans
must protect user privacy and information security during data processing and
storage. Europe stresses citizen data rights and ensures AI adheres to ethical
and legal standards.
4. Ethical considerations and human-AI collaboration:
Fostering AI ethics and achieving effective
human-AI synergy is crucial, requiring attention to human-machine interaction
and ethical awareness in professional development.
5. Responsible usage:
There's a need for strict regulation and
control to prevent AI from being used for unethical or harmful purposes, as the
powerful capabilities of generative AI pose a risk of potential misuse.
What is the role of Generative Artificial Intelligence in Emotional Companionship?
What are the ethical implications of AI Emotional companionship?
The application of AI companions in emotional companionship uses generative AI technology
to create virtual characters or robots that can understand and respond to human
emotions to provide emotional companionship and social support. This is
especially popular among young people.
An example is:
By 2035 the combination of virtual idol Ming
Rui and AI companion Qing Shun, which sparked a global trend. Users like Enron
could set a role for the virtual avatar Minrui and chat daily, with the avatar
learning to become a loyal friend.
•A popular AI
companion software launched in 2017 now has over 10 million users, and approximately 40% of users have
set their relationship with AI to that of a romantic partner.
* The increasing popularity of virtual companions has raised ethical concerns
about whether this virtual companionship will replace real human interaction.
Also, Loneliness and isolation are a real concern
What are the Frontiers in Generative Artificial Intelligence Research?
1. The application of generative AI in the
arts involves machines generating new, original content by learning and
imitating human creative patterns and styles. This is applied in art creation,
music creation, image generation, and natural language generation.
2. Generative AI can be a Creative Assistant, collaborating with human artists
who provide creativity and inspiration.
3. The combination of "Generative AI + Photoshop" is noted as a
"rocket launcher" for promoting human artistic creativity, able to
complete drawing tasks in seconds.
4. In creative performance, AI can generate creative content such as art,
music, and literature. Generative Adversarial Networks (GANs) can generate
realistic images and videos.
What are AI regulations challenges?
The three challenges for AI regulation were proposed by Tom Wheeler and are: speed
issues, content of regulation, and supervisors and manner of regulation.
1. The speed problem is the first challenge:
•Unlike the standardization of the industrial era, tech companies
now use agile management that is transparent, collaborative, and responsive.
•Governments must learn to be agile in developing regulatory
policies that protect consumers' rights without hindering innovation and
development.
An example:
Is an
intelligent robot for the elderly that had unintended consequences (stealing
photos, misleading users) due to inadequate regulation.
2. Regulatory content is the second challenge.
3. Regulators and how they are regulated is the third challenge.
According to the authors, ‘’ It is proposed that the government establish a body specifically
responsible for AI regulation, which would use agile oversight to quickly
respond to emergencies’’
Why is AI governance important?
AI governance is important for several reasons:
1. it helps protect the rights of data subjects ensures transparency in data
sources, maintains accuracy, and identifies responsible parties.
2. It helps ensure that the application of AI technology does not pose
uncontrollable risks to society.
For instance, the Artificial Intelligence Act passed by the European
Parliament includes strict prohibitions on AI systems that pose unacceptable
risks to human safety (e.g., systems that purposefully manipulate technology).
3. It can facilitate cooperation among nations, establish global ethical
guidelines and standards for AI, and promote the sustainable development of AI.
According to the Authors’ ‘‘Both the United States and China have
implemented steps in AI governance.’’
Why does Open-source matter for Generative
AI Business Models?
1. The open source model involves open source code that allows others to use,
copy, modify, and redistribute the source code, and innovate, optimize, and
iterate on it.
2. The advantage of the open source model is that it attracts the attention and
participation of many developers, and promotes technology sharing and
cooperation. Its disadvantage is that commercial gains are relatively limited.
3. The licensing model attracts partners and obtains rich benefits but requires
continuously maintaining the licensing cooperation relationship.
4. The history of open source is a constant struggle and balancing act between
the freedom of software innovation and the benefits of copyright.
5. The government is encouraged to get technology companies to join the open
source ecosystem to help the development of China's ChatGPT model.
6. The big model of artificial intelligence is considered not only human tools
and assistants, but also the new operating system of the artificial
intelligence era.
What is the next Generative AI
Race?
The race for a
carbon-based + silicon-based world: brain-computer interfaces to look forward
to
A. Brain-computer interface technology :
1. Relates to the direct interaction of the human brain with a computer or external device, potentially allowing people to control computers through pure ideas, without the need for any external equipment.
2. In the medical field, it holds promise, exemplified by a woman who regained
the ability to stand and walk through this technology after being incapacitated
by a rare nervous system disorder.
3.Brain-computer interface technology makes generative AI more interesting by
allowing generative artificial intelligence to interact with and control the
human brain. This is helpful to understand the human brain's information
processing mechanism and apply it to generative AI development.
4. Liu Jia proposed that "General artificial intelligence will be as
important as hydropower in the future," affecting skill-acquisition, child
development, and the direction of human development, potentially leading to the
advent of a second cognitive revolution.
5. In the near future, humans will enter a "carbon-based +
silicon-based" world, with brain-computer interface technology becoming
the most high-profile innovation.
B.Generative Artificial Intelligence is Becoming Smarter
1.AI's self-learning and evolution is a process involving techniques like human
feedback training and reinforcement learning.
2. Through continuous interaction with humans and receiving human feedback, AI
systems learn to better adapt to human needs.
3. Reinforcement learning involves continuous trial and error and feedback to
learn optimal decision-making strategies, with significant results in areas
like autonomous driving and robotic control.
4. With continuous advancements, it's possible we could create digital people
with a "soul" in the future.
How does Generative Artificial
Intelligence relate to the Capital Markets?
1. The influence of the P plug-in on the capital market is mainly reflected in
the optimization of investment portfolio, automatic decision-making and risk
management.
2. The P plug-in helps investors optimize investment portfolio and improve
investment returns through technical analysis, basic analysis, and asset
allocation strategies.
3.Its automatic decision-making function
eliminates emotional and subjective factors, making investment decisions more
rational.
4. However, over-reliance on the P plug-in may have negative effects.
For example, one investor found that the plug-in sometimes relied too heavily on automated decision-making, resulting in a single investment strategy. Another investor worried that if everyone used the plug-in, the market could become too mechanized and lack human factors, potentially triggering market imbalances.
5. Individual investors facing the generative AI capital market need to possess
the three qualities of rational thinking, flexible response, and prudent
decision-making to fully leverage the advantages of AI plug-ins while
maintaining active control over market fluctuations, achieving more stable
investment returns.
To conclude,the book "AI
Awakening," by Lu Junqun and Li Xuan, provides a professional economic
analysis of Generative AI (AI) and its profound societal impact.
Generative AI's value stems from its powerful capabilities across multiple fields, including business,
education, and the arts. This capability is fundamentally supported by
breakthroughs in brain science, data, algorithms, and computing power. The
emergence of large models, exemplified by the Transformer breakthrough, has
driven innovation and lowered market barriers for startups, despite current
computational and accuracy challenges.
The proliferation of Generative AI necessitates robust governance to mitigate
risks.
Key challenges involve ensuring content
authenticity, reliability, and adherence to moral guidelines to prevent
discrimination or false information. Protecting user data and information
security, especially with applications like AI emotional companions, which
raise concerns about replacing real human interaction.
Moreover, addressing the speed of technological
change by establishing a dedicated government body for AI regulation that uses
agile oversight.
The ultimate competitive race is toward a "carbon-based + silicon-based
world". The most high-profile innovation in this race is Brain-Computer
Interface (BCI) technology. BCI will allow Generative AI to interact directly
with and control the human brain, potentially leading to a second cognitive
revolution and shaping the future direction of human development.
Look forward to your thoughts- What is your AI Awakening Moment?
References:
1.https://hai.stanford.edu/news/ai-awakening
2.https://aiig.tsinghua.edu.cn/
Comments
Post a Comment